corgy-w opened a new issue, #7735:
URL: https://github.com/apache/seatunnel/issues/7735

   ### Search before asking
   
   - [X] I had searched in the 
[issues](https://github.com/apache/seatunnel/issues?q=is%3Aissue+label%3A%22bug%22)
 and found no similar issues.
   
   
   ### What happened
   
   ## Usage
   
   1. In cases where the task does not encounter errors, the amount of data 
reaching the sink represents the number of records inserted into the sink.
   2. In the event of task failures:
      1. Total failure, meaning that no data is written within a single batch 
period; result: written = read.
      2. Partial failure, where data in the stream encounters format issues or 
other problems, leading to write failures. In this case, the amount written is 
N * batchSize; result: written = read.
      3. Note the `job.retry.times` configuration, which defaults to 3. This 
means that after a task failure, the system will retry three times by default, 
resulting in a read count that is four times the original value, with the same 
applying to the write count.
   
   ## Pain Points
   
   1. There is a lack of corresponding metadata available for dependency 
development when executing data reconciliation using metric data.
   2. Troubleshooting magnitude issues is challenging without reliable evidence 
to support the findings.
   
   ### SeaTunnel Version
   
   dev
   
   ### SeaTunnel Config
   
   ```conf
   -
   ```
   
   
   ### Running Command
   
   ```shell
   -
   ```
   
   
   ### Error Exception
   
   ```log
   -
   ```
   
   
   ### Zeta or Flink or Spark Version
   
   _No response_
   
   ### Java or Scala Version
   
   _No response_
   
   ### Screenshots
   
   _No response_
   
   ### Are you willing to submit PR?
   
   - [X] Yes I am willing to submit a PR!
   
   ### Code of Conduct
   
   - [X] I agree to follow this project's [Code of 
Conduct](https://www.apache.org/foundation/policies/conduct)
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to